ERROR CODES & MESSAGES

This Appendix describes the error conditions recognised by this package and the associated codes and messages. If it is necessary to test for any of these error conditions, then symbolic constants (defined by Fortran PARAMETER statements) should be used to identify the error code in question.

The names of PRIMDAT error codes are of the form:

PRM__$<$code$>$

where $<$code$>$ specifies the error condition. Their numerical values are defined in the file PRM_ERR, and may be incorporated into an application with the statement:

      INCLUDE 'PRM_ERR'

The following gives the names of the symbolic constants defined in this file, together with the associated error messages and an explanation of each error condition:

PRM__ARGIN, function argument invalid
An invalid argument was supplied to a mathematical function.

PRM__FLTDZ, floating point divide by zero
An attempt was made to perform floating point division by zero.

PRM__FLTOF, floating point overflow
The absolute value of the result of a floating point calculation or type conversion exceeded the largest floating point number which can be represented on the machine.

PRM__FLTUF, floating point underflow
The absolute value of the result of a floating point calculation or type conversion was smaller than the smallest non-zero floating point number which can be represented on the machine. This error condition will not normally be used on a VAX unless underflow detection has been explicitly enabled.

PRM__INTDZ, integer divide by zero
An attempt was made to perform integer division by zero.

PRM__INTOF, integer overflow
The result of an integer (non-floating point) calculation or type conversion was outside the number range which can be represented by the data type in question.

PRM__LOGZN, logarithm of zero or negative number
An attempt was made to take the logarithm of zero or a negative number.

PRM__SLOST, numerical significance lost
Numerical significance was lost during evaluation of a mathematical function; a result could not be calculated with any meaningful precision.

PRM__SQRNG, square root of negative number
An attempt was made to take the square root of a negative number.

PRM__UDEXP, undefined exponentiation
An attempt was made to raise the number zero to a negative power, or to raise a negative number to a non-integer power.